Search

Run AI coding agents from your phone. Bridges Claude Code, Codex, OpenCode, Pi, Gemini CLI, and Amp to Telegram with interactive permissions, voice input, cost tracking, and live progress.

LangChain-compatible integration for AgentOracle

Harness scaffold tool — initialize AI-assisted development environments for any project type and AI coding tool.

Shared TypeScript type definitions for the Tepa agent pipeline

🍋 Stale content eviction for OpenClaw — squeeze images, tool results, and exec outputs from context. Compaction fires 2-3x less often.

Self-evolving skill engine for OpenClaw agents — Phase 1 core + Phase 2 proactive intelligence + Phase 3 self-validation

WhatsApp AI Bot — Automate, Reply, Schedule, Assist. Powered by OpenAI Agents SDK.

AI agent skills for publishing Gardensale catalog items to marketplaces

ScopeBlind evidence wrapper for LangChain tool calls. Automatically creates Veritas Acta receipts for every tool invocation.

SidClaw governance integration for LangChain.js — policy evaluation, human approval, and audit trails

Spec-driven AI development toolkit — design, evaluate, stress-test, and certify autonomous agents from your terminal

Multi-agent failure detection for production AI systems

Bridge WeChat messages to any terminal AI agent (Claude Code, Codex) via ilink protocol

Lightweight framework for building LLM agents with tool use, memory, and multi-round research pipelines.

loophaus — Control plane for coding agents. Iterative dev loops with multi-agent orchestration.

An ACP-compatible coding agent powered by the Claude Agent SDK with invocation-specific Claude state

CLI that packages, installs, and manages Claude Code subagent personas and workflow knowledge assets.

SQLite backend for db0. Pure JavaScript via sql.js, zero native dependencies. Local-first persistent memory for AI agents.

An LLM agent friendly file scanner

Smart model routing extension for OpenClaw — dynamically selects the best AI model per conversation turn